Reform of the Measurement of All Aspects of Fixed Capital

The national estimates of fixed capital comprise four vital measures: investment, the most dynamic component of GDP; the gross capital stock, which determines national output capacity; the net stock, the value of future capacity and a large part of measured national wealth; and capital consumption, the annual amount by which the latter shrinks, through wear-and-tear and obsolescence. Yet, it is contended here, by an anomaly of national accounting, the means of producing these statistics are flawed. The annual investment survey lacks vital data categories, and the so-called perpetual inventory method (PIM), using assumed values of service lives, estimates the capital stock and capital consumption with gross inaccuracy. A good system of fixed capital statistics is not possible without a correctly designed survey questionnaire. Unfortunately, present questionnaires generally do not adequately capture the logic of capital accumulation, routinely entered in the company accounts. These contain three distinct investment components, capital expenditure on off-the-shelf items directly adding to the producing stock; capital expenditures for work-in-progress that will enter service only in the sometimes distant future; and large account transfers, from time to time, when completed work enters service. Further, most agencies' questionnaires do not include reconciliations, to ensure accuracy and to screen out accounting distortions. The three components can be combined to give two definitions of investment, what might be called the national accounts and capital stock definitions, abbreviated, respectively, 'N' and 'C'. Most agencies now survey only the N definition, capital expenditure, the sum of the first two components. The C definition is the sum of the first and third items, and is the annual addition to the productive or in-service stock. The Netherlands surveys only this definition. While over, say, ten years, the cumulative values of the two definitions are approximately equal, in individual years, one definition may be a small fraction of the other. One consequence of the present defective data base is investment statistics, both as a component of GDP, and for monitoring the economy, that are less accurate and informative than they should be: During the year, how much was added to the stock of in-service capital? Of this, how much was direct off-the-shelf additions, how much fruition of decisions made years ago? How much was spent to create future capacity? Acknowledgments. Without the support over the years of Derek Blades and Terry Gigantes, this work would probably never have been done. They early recognised that capital …